UE Méthodes inverses et assimilation

Diplômes intégrant cet élément pédagogique :

Descriptif

Inverse methods aim at using measurements to infer the values of some unobserved parameters characterizing an observed system. Data assimilation is a specific class of inverse problems where the observed system is dynamical, and measurements are distributed in time. Both approaches are widely used in Earth and climate sciences. Examples are the retrieval of Earth mantle properties from seismic measurements, of the characterization of ocean currents from satellite observations of sea level.

The course aims at introducing theoretical and methodological basics for the resolution of inverse and data assimilation problems in Earth sciences. It is given through interactive lectures including the presentation and manipulation of numerical practices in matlab and python. The content of the course is transverse: the methods introduced can be used in a large variety of applications.

 

 

Pré-requis

Standard notions in linear algebra, differential calculus, statistics.

Compétences visées

To be able to pose an inverse or a data assimilation simple problem et implement a method to solve it. Develop a critical eye on results. To be able to communicate with experts.

Evaluation :
A personal project (50%) and a final examination (3 hours, 50%)

Informations complémentaires

Lieu(x) : Grenoble - Saint-Martin d'Hères
Langue(s) : Francais